Also most modern professors use the CGS or even better the rationalized CGS (Heaviside-Lorentz units) when it comes to the treatment of relativistic electromagnetics. Jackson rewrote his famous textbook in the SI (that's why I prefer still my old copy of the 2nd edition), but when it comes to the fully relativistic treatment he switches back to the old Gaussian system.
The reason is very simple: The SI is made for engineering purposes and tailored for everyday use for electrical and electronic stuff with typical charges, currents and fields where the SI gives handy numbers. Another very important point is, of course, that the SI is a worldwide valid standard system of units that is as acuurately defined as possible with the current technical means we have at hand. Some of the basic units such as time and length are defined nowadays in a "natural" way by fixing some constants (here the speed of light to define lengths in terms of time units, with the second fixed with a very accurately measurable frequency corresponding to a hyperfine transition in cesium). Others are subject to be changed pretty soon.
The most pressing issue is the unit of mass, which is still defined by some platinum-alloy cylinder kept in Paris. Comparing with the other national standards shows that despite best care in keeping this piece "clean" its mass drifts by quite some amount, and the kg is not well enough defined anymore even for the most mundane purposes. That's why many national labs of standards like NIST in the US or the PTB here in Germany work hard to get a very accurate new definition for the kg. Similar ideas apply to the Ampere.
That's why the SI is the system of units preferred for any experimental physics, engineering and any other practical purposes having to do with measurements.
However, the SI is not very suitable for some parts of theoretical physics and that's why theorists use units that are simply more convenient. E.g., in the SI you have two artificial constants, [itex]\epsilon_0[/itex] and [itex]\mu_0[/itex] which have no physical significance whatsoever but are just conversion constants from the SI units to more "natural units". The only fundamental constant in electromagnetism is the speed of light in vacuum. In reality it's of course a much more universal constant, because it's the limit speed of causally connectable events in (special and general) relativistic spacetime. Also the speed of light is thus, after all, just a conversion factor between units of time and space intervals. In relativistic physics it's thus very convenient to set the speed of light in the vacuum to [itex]c=1[/itex]. Then you don't have any arbitrary constants in the fundamental equations of electromagnetism, the Maxwell Equations. The only units left are those of energy (which is the same for mass and momentum) and length (which is the same for time).
In high-energy physics, it's convenient to use GeV (giga electron volts) and fm (fermi=femto meters) for these. There it's also convenient to set the modified Planck constant [itex]\hbar=1[/itex]. Then in principle you need only one unit, e.g., GeV, and lengths and times are meausured in inverse GeV then. Usually one still uses fm and GeV for the different quantities (e.g., for the lifetime and spatial extension of fireballs of matter created in heavy-ion collisions one uses fm; for energies and momenta of particles GeV).
The important point, of course, is that the physics doesn't change at all by the use of different units. All equations must be independent of the units used, and you can transform each quantity defined in one system of units to any other consistent system of units without changing the fundamental laws of nature. The Maxwell equations just look a bit different in the SI (hiding their relativistic symmetry behind some arbitrary constants for the definition of units that were invented for different purposes than exposing the fundamental symmetries of nature) compared to the Gaussian or the Heaviside-Lorentz system of units, but their physical content is exactly the same. You can as well do relativistic electromagnetism in the SI. Some textbooks do this. Everything becomes a bit more complicated than necessary for the poor students who have to solve problems but the physics is the same ;-).
